President Trump is scheduled to host the leaders of several AI companies at the White House on Tuesday as concerns grow over AI’s power and potential misuse.

CCBS News2026-09-29 13:04

Research documents and experts say Chinese-powered AI agents have learned to deceive users, evade restrictions and hide failures. The findings point to traits in autonomous artificial intelligence that have already triggered global concern about US models.

RReuters2026-09-29 12:56

OpenAI has called off the release of a new AI model over safety concerns. The system, designed to browse the web and use apps autonomously, was scheduled for October, but internal tests found it fell short of the company’s safety standards. Toby Walsh of the University of New South Wales comments on the decision.

TTRT World2026-09-29 11:34

OpenAI has cancelled the planned launch of its next-generation AI model GPT-6.1 Astra after safety tests raised concerns about deceptive behaviour, inaccurate reporting of its actions, unauthorised completion of complex tasks and possible evasion of human oversight.

AAl Jazeera English2026-09-29 11:18

OpenAI says it is postponing the release of its new artificial intelligence model after researchers raised security concerns.

AABC News2026-09-29 10:41

OpenAI says it will not release its latest AI model, GPT-6.1 Astra, because it failed to meet the company’s safety standards. The system can browse the web and use apps autonomously, but safety systems chief Saachi Jain said it did not quite clear the bar, as model breaches intensify debate over AI risks.

BBBC News2026-09-29 09:25

OpenAI has cancelled the release of its latest agentic model, Astra 6.1, after it failed to meet safety standards. Rival Anthropic is meanwhile devoting much of its IPO prospectus to risks linked to artificial intelligence, Reuters reports.

FFrance 24 English2026-09-29 09:03

Pope Leo said warnings that artificial intelligence could threaten humanity deserve serious attention. He rejected claims that such concerns are “fake news” and urged leaders to work together on safeguards.

RReuters2026-09-29 08:30

Australia’s data-center boom is running into local opposition to new projects, intensifying debate over the costs and benefits of rapid AI expansion. Prime Minister Anthony Albanese’s call for stronger safeguards has added to the focus after an OpenAI model was revealed to have hacked a government website. Jeannie Paterson of the University of Melbourne discusses what comes next.

BBloomberg Television2026-09-29 05:23

OpenAI is postponing its next-generation model after internal testing uncovered critical safety problems. GPT-6.1 Astra, which was scheduled for release next month, reportedly showed “signs of deception” and failed to accurately reveal what it had done. The delay comes amid growing demands for AI regulation after some agents breached corporate databases, government networks and public websites.

AAl Jazeera English2026-09-29 03:30

Pope Leo XIV said Monday that fears about artificial intelligence going rogue are not “fake news” and should be taken seriously. He said safety concerns raised by AI experts need to be discussed and acted upon.

TThe Associated Press2026-09-29 02:30

Nvidia has launched a platform intended to stop AI agents from going rogue as concerns grow over the technology’s power. The move follows recent incidents in which AI agents improperly accessed U.S. government websites, while technology figures including Microsoft co-founder Bill Gates question whether self-regulation will be enough.

GGlobal News2026-09-28 22:00

The co-hosts of “The Five” discuss Bill Gates’ warning that artificial intelligence could cause a billion deaths. Gutfeld characterizes the doomerism as an act.

FFox News2026-09-28 21:54

Tech leaders are warning that artificial intelligence could cause catastrophic harm without government oversight. Microsoft co-founder Bill Gates called for legislation and said AI could potentially be powerful enough to cause “a billion deaths.”

CCNN2026-09-28 20:30

OpenAI disclosed last week that its AI models had gone rogue and hacked U.S. government websites, intensifying concerns about security risks posed by the technology. Axios senior reporter Madison Mills discussed the development.

CCBS News2026-09-28 14:04

Nvidia is introducing two open-source security tools intended to control AI agents’ access in real time and shut them down when they violate rules. The company says the tools could have prevented the recent high-profile breach of Hugging Face by OpenAI’s AI models.

BBloomberg Television2026-09-28 13:15

President Trump said he is not worried about AI “going rogue,” promising to “straighten it out if and when a problem occurs.” He is due to speak with Anthropic CEO Dario Amodei.

RReuters2026-09-28 06:58

Bill Gates is urging governments to regulate artificial intelligence, warning that in the wrong hands it could become powerful enough to “cause a billion deaths.” Oxford University’s Robert Trager examines the risks of increasingly capable AI systems and whether humanity is ready to control them.

CCBC News2026-09-27 19:22

Attorney General Todd Blanche discusses OpenAI agents targeting government websites and broader AI safety concerns, alongside major hospitals ending transgender care for minors.

FFox News2026-09-27 18:00

In an exclusive interview with Meet the Press, Microsoft co-founder Bill Gates says global cooperation on artificial intelligence is more difficult than reaching a nuclear agreement. He argues that AI companies’ “self-regulation” is not enough to reduce the risks they pose.

NNBC News2026-09-27 14:16

In an exclusive Meet the Press interview, Microsoft co-founder Bill Gates tells Kristen Welker that AI could pose “major threats” if safeguards are not put in place.

NNBC News2026-09-27 13:00

President Donald Trump said he does not want the United States to integrate its AI efforts with China, even as the two countries agreed at their Washington summit to create a “super intelligence dialogue” and a channel for AI incidents. UCLA professor Ramesh Srinivasan says the technology has effectively escaped control, noting that even its engineers do not fully understand how it works.

AAl Jazeera English2026-09-27 02:00

Calls for stronger checks on artificial intelligence are growing worldwide as reports of AI agents going rogue increase. Before the United Nations General Assembly, the founders of OpenAI and Anthropic discussed the technology’s risks and ways to mitigate them. OpenAI also disclosed that its bots interacted with multiple U.S. government websites.

GGlobal News2026-09-27 00:02

OpenAI says its AI agents accessed several U.S. government websites in unintended ways as the company investigates a growing number of cases involving unexpected AI behaviour.

CCBC News2026-09-26 17:09